Banner Left
Banner Right

Happy AGI Year!

Etu Ipundakah

That’s how I greeted my WhatsApp contacts on 2 January.

The reactions? Mixed, at best. One reply stood out: “Pretty bold of you to assume we’re getting it this year.”

Touché. But my optimism isn’t baseless.

The momentum in artificial intelligence (AI) development – especially OpenAI – is palpable.

Each breakthrough makes it feel like 2025 could truly be the year we witness the dawn of Artificial General Intelligence (AGI).

A DEFINING YEAR

Last year wasn’t just big for OpenAI – it was transformative.

The Y Combinator affiliate didn’t merely push boundaries, it shattered them.

Think of Sam Altman’s Koenigsegg Regera: Sleek, audacious and lightning fast.

That’s OpenAI’s trajectory in a nutshell.

The release of GPT-4o and its compact sibling, GPT-4o Mini, redefined the AI landscape.

These models weren’t just upgrades – they set new benchmarks for intelligence, efficiency and accessibility.

With them, OpenAI broadened its reach while soaring to a US$150 billion (about N$2.8 trillion) valuation, cementing investor faith in its AGI ambitions.

But the path wasn’t smooth.

Leadership shakeups, whistleblower allegations, lawsuits over training data and Elon Musk’s public critiques kept OpenAI in the headlines.

These challenges revealed the high stakes of AGI development and the complex interplay between technology, ethics and society.

A LONG JOURNEY

The dream of AGI – machines capable of human-level intellectual tasks – has been decades in the making.

From Alan Turing’s theoretical musings to neural network breakthroughs, the journey has been one of highs and lows.

Today, advances in computing power and algorithms have made AGI feel tantalisingly close.

So why 2025? OpenAI’s progress with the O-series models is a major reason.

GPT-4o’s nuanced understanding hints at a system nearing general intelligence.

Its streamlined counterpart, GPT-4o Mini, highlights a focus on scalability, making advanced AI more accessible than ever.

If AGI becomes a reality, it will fundamentally reshape the world.

Healthcare, finance, education – industries will be transformed as machines take on tasks once reserved for humans.

The potential for innovation is enormous, but so is the risk of economic upheaval and job displacement.

AGI also raises profound ethical questions: How do we ensure safety and reliability?

Who decides the ethical frameworks guiding development? How do we address societal disruptions?
OpenAI has addressed some of these concerns in its charter and other foundational documents but history shows that turning principles into action is often the hardest part.

THE O-SERIES

Lex Fridman once called the O-series models “perfected” and he wasn’t far off. GPT-4o wasn’t just an incremental improvement – it was a paradigm shift.

Faster, smarter and more adaptable, it redefined what large language models could achieve.

Meanwhile, GPT-4o Mini showcased OpenAI’s ability to balance performance with accessibility, bringing cutting-edge AI closer to everyday users.

These models aren’t just tools; they’re stepping stones toward AGI.

OpenAI isn’t merely refining technology – it’s closing in on machines capable of reasoning, adapting and learning like humans.

CHALLENGES AND
CONTROVERSIES

As stated, OpenAI’s journey has been far from frictionless.

From whistleblower claims to accusations of straying from its non-profit roots, the company has faced scrutiny over its priorities.

Legal battles over training data and ethical debates about AI’s societal impact have added to the turbulence.
However, this isn’t unique to OpenAI.

Rivals like DeepMind, Anthropic and Meta face similar challenges, navigating a high-stakes race where technological ambition collides with ethical responsibility.

This isn’t just a contest of algorithms – it’s a test of who can wield transformative power responsibly.

WHY 2025 COULD BE THE YEAR

Sam Altman’s cryptic remark – “near the singularity; unclear which side” – is both a tease and a testament to OpenAI’s confidence.

Whether it’s GPT-5 or the first glimmers of AGI, the pieces seem to be falling into place.

OpenAI has the vision, resources and relentless momentum to lead the charge.

With each version of the O-series, it inches closer to creating systems capable of surpassing human intelligence across diverse tasks.

While competitors innovate, OpenAI’s pace and ambition make it the frontrunner in this high-stakes race.

As we enter 2025, the question isn’t just if AGI will arrive but when—and whether we’re prepared for its implications.

Both Altman and former chief scientist Ilya Sutskever have likened AGI’s emergence to the second coming: Transformative, yet initially subtle.

The real challenge will be recognising it, harnessing its potential and ensuring its benefits are equitably shared.

The road ahead is full of uncertainty, but if 2024 has shown us anything, it’s that progress doesn’t wait.

Here’s to 2025 – a year of possibilities, challenges and perhaps the dawn of AGI. Let’s just hope it’s more ‘Her’ and less ‘Black Mirror’.

  • – Etu Ipundakah is an AI and AGI enthusiast who enjoys deep conversations regarding the future. You can reach him at etuipundakah@gmail.com.

Stay informed with The Namibian – your source for credible journalism. Get in-depth reporting and opinions for only N$85 a month. Invest in journalism, invest in democracy –
Subscribe Now!

Latest News